NVIDIA B100 vs NVIDIA GeForce RTX 4070 Ti SUPER
We compared two Desktop platform GPUs: 96GB VRAM B100 and 16GB VRAM GeForce RTX 4070 Ti SUPER to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA B100 's Advantages
Released 10 months late
More VRAM (96GB vs 16GB)
Larger VRAM bandwidth (4.10TB/s vs 672.3GB/s)
8448 additional rendering cores
NVIDIA GeForce RTX 4070 Ti SUPER 's Advantages
Boost Clock has increased by 42% (2610MHz vs 1837MHz)
Lower TDP (285W vs 1000W)
Score
Benchmark
FP32 (float)
B100
+40%
62.08 TFLOPS
GeForce RTX 4070 Ti SUPER
44.1 TFLOPS
Graphics Card
Nov 2024
Release Date
Jan 2024
Server Blackwell
Generation
GeForce 40
Desktop
Type
Desktop
PCIe 5.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1665 MHz
Base Clock
2340 MHz
1837 MHz
Boost Clock
2610 MHz
2000 MHz
Memory Clock
1313 MHz
Memory
96GB
Memory Size
16GB
HBM3e
Memory Type
GDDR6X
4096bit
Memory Bus
256bit
4.10TB/s
Bandwidth
672.3GB/s
Render Config
-
-
-
132
SM Count
66
16896
Shading Units
8448
528
TMUs
264
24
ROPs
96
528
Tensor Cores
264
-
RT Cores
66
256 KB (per SM)
L1 Cache
128 KB (per SM)
50 MB
L2 Cache
48 MB
-
-
-
Theoretical Performance
44.09 GPixel/s
Pixel Rate
250.6 GPixel/s
969.9 GTexel/s
Texture Rate
689.0 GTexel/s
248.3 TFLOPS
FP16 (half)
44.10 TFLOPS
62.08 TFLOPS
FP32 (float)
44.10 TFLOPS
31.04 TFLOPS
FP64 (double)
689.0 GFLOPS
Board Design
1000W
TDP
285W
1400 W
Suggested PSU
600 W
No outputs
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
-
Power Connectors
1x 16-pin
Graphics Processor
GB102
GPU Name
AD103
-
GPU Variant
AD103-275-A1
Blackwell
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
104 billion
Transistors
45.9 billion
Unknown
Die Size
379 mm²
Graphics Features
N/A
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
10.1
CUDA
8.9
N/A
Shader Model
6.7
